Talent rating system, talent rating method, talent rating device, and talent rating program

Abstract

One embodiment of the present disclosure is a talent rating system provided with: a subject terminal device; an external device for evaluating the subject; and a talent rating device to which the subject terminal device and the external device are connected via a communication network, the talent rating device being provided with: a calculation unit for inputting talent information of the subject acquired from the subject terminal device and an evaluation result of the subject acquired from the external device; outputting feature information of the object person according to the talent information and the evaluation result; and a rating determination unit that determines the rating of the subject on the basis of the feature information output from the calculation unit.

Description

Technical Field

[0001] The present disclosure relates to a talent rating system, a talent rating method, a talent rating device and a talent rating program. Background Art

[0002] Conventionally, as a technology related to personnel auction, for example, a personnel auction system is known as Patent Document 1. The personnel auction system described in Patent Document 1 uses qualification certificates provided by qualification institutions as objective ability information of job seekers to match the needs of both job seekers and recruiting companies.

[0003] Prior art literature

[0004] Patent Document 1: Japanese Patent Application Publication No. 2015-69507 Summary of the invention

[0005] However, in the talent auction described in Patent Document 1, it is most ideal to conduct a comprehensive evaluation that takes into account other factors, such as career history and evaluation from the workplace, in addition to the qualifications of job seekers.

[0006] The present disclosure has been made in view of such circumstances, and an object of the present disclosure is to provide a talent rating system, a talent rating method, a talent rating device, and a talent rating program that can comprehensively evaluate talents.

[0007] The present disclosure is completed to solve the above-mentioned problems. One embodiment of the present disclosure is a talent rating system, comprising: a subject terminal device used by a subject; an external device for evaluating the subject; and a talent rating device, to which the subject terminal device and the external device are connected via a communication network, and the talent rating device comprises: a calculation unit, which inputs talent information of the subject obtained from the subject terminal device and an evaluation result of the subject obtained from the external device, and outputs characteristic information of the subject based on the talent information and the evaluation result; and a rating determination unit, which determines the rating of the subject based on the characteristic information output from the calculation unit.

[0008] Another aspect of the present disclosure is a talent rating method of a talent rating system, the talent rating system comprising: a subject terminal device used by a subject; an external device for evaluating the subject; and a talent rating device, the subject terminal device and the external device being connected via a communication network, wherein the talent rating method comprises: a step in which the subject terminal device sends talent information of the subject to the talent rating device; a step in which the external device sends an evaluation result of the subject to the talent rating device; a step in which the talent rating device obtains the talent information of the subject and the evaluation result of the subject; a step in which the talent rating device calculates characteristic information of the subject based on the talent information and the evaluation result; and a step in which the talent rating device determines a rating of the subject based on the characteristic information.

[0009] Another aspect of the present disclosure is a talent rating device, which is connected to a subject terminal device used by a subject and an external device for evaluating the subject via a communication network, wherein the talent rating device comprises: a calculation unit, which inputs talent information of the subject obtained from the subject terminal device and an evaluation result of the subject obtained from the external device, and outputs characteristic information of the subject based on the talent information and the evaluation result; and a rating determination unit, which determines the rating of the subject based on the characteristic information output from the calculation unit.

[0010] Another aspect of the present disclosure is a talent rating program, which enables a computer of a talent rating device connected to a subject terminal device used by a subject and an external device that evaluates the subject via a communication network to function as the following components: a calculation unit, which inputs talent information of the subject obtained from the subject terminal device and an evaluation result of the subject obtained from the external device, and outputs characteristic information of the subject based on the talent information and the evaluation result; and a rating determination unit, which determines the rating of the subject based on the characteristic information output from the calculation unit.

[0011] According to the present disclosure, it is possible to comprehensively evaluate talents. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 : is a block diagram showing an example of a talent rating system in an embodiment.

[0013] Figure 2 This is a block diagram showing an example of the configuration of a talent bidding device and a talent rating device in the embodiment.

[0014] Figure 3 This is a block diagram showing an example of a specific structure of the talent rating system in the embodiment.

[0015] Figure 4 This is a diagram showing an example of a multivariate analysis unit in the embodiment.

[0016] Figure 5 This is a diagram showing an example of a feature vector mapping process in the embodiment.

[0017] Figure 6 This is a diagram showing an example of a subject person extraction process using a feature vector in the embodiment.

[0018] Figure 7 This is a diagram showing an example of a bidding calculation process in the embodiment.

[0019] Figure 8 This is a sequence diagram showing an example of the operation process of the talent rating system in the embodiment. DETAILED DESCRIPTION

[0020] Hereinafter, a talent rating system, a talent rating method, a talent rating device, and a talent rating program to which the present invention is applied will be described with reference to the accompanying drawings.

[0021] Figure 1 1 is a block diagram showing an example of a talent rating system 1 in an embodiment. The talent rating system 1 includes, for example, a talent bidding device 100, a talent rating device 200, a subject terminal device 300, a dispatch source terminal device 400, a dispatch destination terminal device 500, an external device 600, and a training content providing device 700. The talent bidding device 100, the talent rating device 200, the subject terminal device 300, the dispatch source terminal device 400, the dispatch destination terminal device 500, the external device 600, and the training content providing device 700 are connected via a communication network NW, and have a communication interface (not shown) such as a NIC (Network Interface Card) or a wireless communication module for connecting to a network such as the Internet. The network may include a general network such as the Internet, and a dedicated network such as a local 5G or WiFi (registered trademark in Japan).

[0022] The subject terminal device 300 is an information processing device such as a smartphone or a personal computer used by the subject. The dispatch source terminal device 400 is an information processing device such as a smartphone or a personal computer used by the dispatch source merchant. The dispatch destination terminal device 500 is an information processing device such as a smartphone or a personal computer used by the dispatch destination merchant. The external device 600 is an information processing device such as a server device that receives a request from the subject terminal device 300, information from the dispatch source terminal device 400, and information from the dispatch destination terminal device 500 and evaluates the subject from the outside. The training content providing device 700 is an information processing device such as a server device that receives a request from the subject terminal device 300 and sends content for training and education to the subject terminal device 300.

[0023] Figure 2 1 is a block diagram showing an example of the structure of the talent bidding device 100 and the talent rating device 200 in the embodiment. The talent bidding device 100 is, for example, an information processing device that provides the following services: registering talent information, evaluation results, dispatch source merchant information, and dispatch destination merchant information, and determining the dispatch destination of the target person based on the bid of the dispatch destination merchant. The target person as a talent is, for example, a full-time person with professional qualifications and skills such as a licensed pharmacist. The talent information is information reported by the target person. The talent information may be information including the target person's age, available work location, work style, professional resume, or skills. The evaluation result includes evaluation information such as the test result received by the target person using an external device for evaluating the target person. The evaluation result may be information including the training lecture record of the target person, the labor record at the dispatch destination, or the evaluation of the target person obtained from the dispatch destination terminal device 500. The dispatch source merchant information is information that specifies the dispatch source merchant of the target person. The dispatch destination merchant information is information that specifies the dispatch destination merchant.

[0024] The talent bidding device 100 includes, for example, an issuing unit 110, an information providing unit 120, a bid accepting unit 130, and a dispatch destination determining unit 140. The issuing unit 110, the information providing unit 120, the bid accepting unit 130, and the dispatch destination determining unit 140 are implemented by, for example, a processor such as a CPU (Central Processing Unit) executing a program stored in a program memory.

[0025] The issuing unit 110 issues bid points to the dispatch destination merchant. The issuing unit 110 can issue bid points to the dispatch destination merchant and the dispatch source merchant. Bid points are information that functions as virtual currency and is virtually traded in the talent bidding device 100, the dispatch source merchant, and the dispatch destination merchant for bidding. The issuing unit 110 can perform at least one of a process of periodically issuing bid points based on the registered dispatch destination merchant information, a process of periodically issuing bid points based on the registered dispatch source merchant information, and a process of issuing bid points based on the browsing history of specific content. The specific content is, for example, content including the target person, the dispatch source merchant, and information related to the dispatch source merchant provided by the talent bidding device 100. The specific content can be content displayed when the dispatch destination merchant inputs recruitment conditions, content displayed when the dispatch destination merchant bids, and content displayed when the dispatch source merchant registers the target person. The dispatch destination merchant can own the issued bid points and bid using the owned bid points.

[0026] The information providing unit 120 transmits the talent information obtained from the subject terminal device 300 and the evaluation result obtained from the external device 600 to the dispatch destination terminal device 500. The information providing unit 120 receives, for example, a request including recruitment conditions from the dispatch destination terminal device 500, and returns the talent information and evaluation result of the subject extracted based on the recruitment conditions to the dispatch destination terminal device 500. The recruitment conditions are, for example, the conditions of the talent desired by the dispatch destination merchant in the talent information and the evaluation result.

[0027] The bid accepting unit 130 accepts a bid request including the subject information and the bid point information from the dispatch destination terminal device 500. The subject information is information for identifying the subject. The bid point information is information indicating the bid points used in the bid of the subject among the bid points owned by the dispatch destination merchant.

[0028] The dispatch destination determination unit 140 determines the dispatch destination of the subject person based on the bid request. When the dispatch destination determination unit 140 accepts one bid request for one subject person, it determines the dispatch destination merchant of the dispatch destination terminal device 500 that has sent the bid request as the successful bidder. When the dispatch destination determination unit 140 accepts multiple bid requests for one subject person, it determines the dispatch destination merchant with the largest bid points among the dispatch destination merchants of the dispatch destination terminal device 500 that has sent the bid request as the successful bidder.

[0029] The talent rating device 200 includes, for example, a calculation unit 210, a rating determination unit 220, and a bid determination unit 230. The calculation unit 210, the rating determination unit 220, and the bid determination unit 230 are implemented by, for example, a processor such as a CPU executing a program stored in a program memory.

[0030] The operation unit 210 inputs the talent information of the subject obtained from the subject terminal device 300 and the evaluation result of the subject obtained from the external device 600 as a talent evaluation device, and outputs the feature information of the subject. The feature information is, for example, a high-dimensional feature vector including the talent information of the subject and the evaluation result. The rating determination unit 220 determines the rating of the subject based on the feature information output from the operation unit 210. The rating of the subject is information indicating the value of the subject, such as the demand degree, recommendation degree, and ranking of the subject among a large number of subjects.

[0031] The bid determination unit 230 determines the initial value of the bid of the object person based on the rating determined by the rating determination unit 220. The bid of the object person indicates the number of bid points required for the dispatch destination merchant to bid for the object person. The higher the rating, the higher the initial value of the bid of the object person is set. The bid determination unit 230 makes corrections in such a way that the bid increases as more bid requests are made. The bid determination unit 230 may make corrections in such a way that the bid decreases when there are fewer bid requests. The information provision unit 120 transmits the object person information and the bid determined by the bid determination unit 230 to the dispatch destination terminal device 500 in response to a request from the dispatch destination terminal device 500.

[0032] The dispatch destination determination unit 140 determines a dispatch destination merchant who has won the bid for the target person based on the bid and the bid request determined by the bid determination unit 230 .

[0033] Figure 3 This is a block diagram showing an example of a specific configuration of the talent rating system 1 in the embodiment.

[0034] The talent rating system 1 has a talent matching function and a talent rating function. The talent matching function is realized by, for example, the point management device 102, the subject terminal device 300 as a terminal device operated by the user of the talent rating system 1, the dispatch source terminal device 400 and the dispatch destination terminal device 500, and the talent verification device 104. The point management device 102 and the talent verification device 104 are functional parts realized by the talent bidding device 100.

[0035] The talent rating function includes, for example, an external device 600, a talent registration device 202, a storage device 212, and a learning device 214. The learning device 214 includes, for example, a talent characteristic evaluation unit 214a, a talent value evaluation unit 214b, and a learning information update unit 214c. The talent registration device 202, the storage device 212, and the learning device 214 are functional units implemented by the talent rating device 200.

[0036] The human resource checking device 104 is, for example, a functional unit realized by the human resource bidding device 100. The human resource checking device 104 includes, for example, an information providing unit 120, a bid accepting unit 130, and a dispatch destination determining unit 140.

[0037] The point management device 102 includes an issuing unit 110 and a consuming unit 112. The consuming unit 112 consumes the bidding points by subtracting the bidding points from the bidding points of the successful bidder when the dispatch destination is determined by the dispatch destination determining unit 140. The consuming unit 112 does not subtract the bidding points of the dispatch source merchant other than the successful bidder.

[0038] The human resource registration device 202 includes, for example, a company demand registration unit 204 and a human resource information registration unit 206. The company demand registration unit 204 obtains the candidate information of the candidate who can be dispatched from the dispatch source terminal device 400. The company demand registration unit 204 obtains information indicating the recruitment conditions from the dispatch destination terminal device 500. The company demand registration unit 204 stores the information obtained from the dispatch source terminal device 400 and the dispatch destination terminal device 500 in the storage device 212, thereby registering the information indicating the needs of the dispatch source business and the dispatch destination business. The human resource information registration unit 206 stores the human resource information obtained from the candidate terminal device 300 in the storage device 212.

[0039] The external device 600 is an information processing device that performs a process of evaluating the subject. The external device 600 includes, for example, a test implementation unit 710 and a comprehensive evaluation unit 410. The test implementation unit 710 transmits the test content to the subject terminal device 300, and calculates the test result based on the answer obtained from the subject terminal device 300. The test implementation unit 710 stores the test result in the member information database 212a. The comprehensive evaluation unit 410 obtains, for example, the evaluation information of the subject in the dispatch source terminal device 400 and the viewing (listening) history of the training content provided to the subject from the training content providing device 700, and stores it in the member information database 212a as information for comprehensively evaluating the subject.

[0040] The storage device 212 is an information processing device that stores various information. The storage device 212 includes, for example, a member information database 212a and a learning database 212b. The member information database 212a stores information about the dispatch source merchant, the dispatch destination merchant, and the subject. Specifically, the member information database 212a stores the subject information of the dispatch source merchant, the information indicating the recruitment conditions of the dispatch destination merchant, and the information indicating the talent information and evaluation results of the subject. The learning database 212b obtains the feature vector of the subject and the feature vector indicating the recruitment conditions, and stores the feature vector as learning data.

[0041] The talent characteristic evaluation unit 214a includes, for example, a calculation unit 210. The calculation unit 210 obtains the talent information and evaluation results of the subject from the member information database 212a, and transforms the obtained information into a high-dimensional feature vector. The feature vector of the subject is information that expresses the subject using a high-dimensional space. The calculation unit 210 stores the feature vector of the subject in the learning database 212b. In addition, the talent characteristic evaluation unit 214a transforms the recruitment conditions obtained from the dispatch destination merchant into a feature vector of the recruitment conditions having values ​​corresponding to the expectations of the talent, working hours, location, etc. required by the dispatch destination merchant. The feature vector of the recruitment conditions is information that expresses the recruitment conditions using a high-dimensional space. The calculation unit 210 stores the feature vector of the recruitment conditions in the learning database 212b.

[0042] The talent value evaluation unit 214b includes, for example, a rating determination unit 220 and a bid determination unit 230. The rating determination unit 220 changes the rating of the subject according to the feature vector of the subject obtained from the talent characteristic evaluation unit 214a. The bid determination unit 230 determines the bid of the subject according to the changed rating. The talent value evaluation unit 214b outputs an instruction to change the test content to the test implementation unit 710, and updates the feature vector through the calculation unit 210 according to the test result of the subject who took the changed test, and updates the rating information indicating the rating of the subject and the bid information indicating the bid. The talent value evaluation unit 214b updates the feature vector according to the comprehensive evaluation of the test result and the information of the subject, thereby evaluating the value of the subject.

[0043] The learning information updating unit 214c obtains the feature vector from the human resource evaluation unit 214b and outputs it to the human resource verification device 104. The human resource verification device 104 updates the feature vector of the subject stored in the learning database 212b using the feature vector obtained from the learning information updating unit 214c.

[0044] Figure 41 is a diagram showing an example of a multivariate analysis unit 200A in the embodiment. The multivariate analysis unit 200A performs calculations of feature vectors, ratings, and bids in the talent feature evaluation unit 214a (calculation unit 210) and the talent value evaluation unit 214b (rating determination unit 220, bid determination unit 230). The multivariate analysis unit 200A outputs a feature vector by, for example, inputting self-declared information and external evaluation information and performing multivariate analysis.

[0045] Self-declared information is, for example, information on national qualifications, professional resume information, university major information, professional field and field of expertise information, workable area information, workable time information, work method information such as commuting or remote work, language information, and home information, etc., which are declared by the subject. External evaluation information is, for example, information obtained by external organizations evaluating the subject such as online test information, comprehensive evaluation information of the dispatch destination, and information related to the professional industry to which the subject belongs (such as the Association of Licensed Pharmacists). External organizations include organizers other than the subject such as dispatch source merchants, test organizers, training organizers, and educational content providers. The multivariate analysis unit 200A performs processing using any analysis of logistic regression analysis, comparative risk analysis, variance analysis, multiple regression analysis, discriminant analysis, principal component analysis, factor analysis, or cluster analysis as multivariate analysis. Multivariate analysis can include, for example, inductive processing such as principal component analysis or factor analysis, classification processing such as cluster analysis or discriminant analysis, and predictive processing such as regression analysis.

[0046] Figure 5 This is a diagram showing an example of a feature vector mapping process in the embodiment.

[0047] The operation unit 210 can compress the dimension of the feature vector of the subject and the feature vector of the recruitment condition, and extract the subject having a feature vector close to the feature vector of the recruitment condition according to the distance between the feature vector of the subject and the feature vector of the recruitment condition in the compressed dimension. Figure 5 The left image) is distilled and mapped to 2D space ( Figure 5The method of multivariate analysis 200A (right figure) is known as t-SNE (t-distributed Stochastic Neighbor Embedding). The multivariate analysis unit 200A needs to process feature vectors in real time, but in t-SNE, when comparing feature vectors in a high-dimensional feature vector space, the more input variables, the more computing time is required. Therefore, the multivariate analysis unit 200A can compress the dimension of the feature vector by using t-SNE and RP (Random Projection) and calculate the distance in 2D space, thereby speeding up the comparison processing of the feature vector. For example, the high-dimensional feature vector of the subject is standardized to the 2D space, and the feature vector of the talent required by the dispatch source merchant is also standardized to the 2D space. As a result, the multivariate analysis unit 200A can reduce the subjects who are similar to the talent required by the dispatch source merchant in a short time.

[0048] Figure 6 This is a diagram showing an example of a subject person extraction process using a feature vector in the embodiment.

[0049] The multivariate analysis unit 200A performs multivariate analysis on the information including the recruitment conditions, thereby mapping the feature vector of the recruitment conditions to a high-dimensional feature vector space. When the feature vector of the recruitment conditions converges to an arbitrary point (× in the figure) in the feature vector space, the multivariate analysis unit 200A uses the k-nearest neighbors method to extract k candidates in the order of the subjects with feature vectors that are closest to the arbitrary point. The talent characteristic evaluation unit 214a Figure 6 When the feature vectors of the recruitment conditions are scattered in any region of the feature vector space like the ellipse in FIG. 1 , all candidates included in the feature vector space including the recruitment conditions are extracted. In order to speed up the calculation, the multivariate analysis unit 200A may perform t-SNE or RP to reduce the dimension of the feature vector space, and extract candidates in the reduced feature vector space.

[0050] The multivariate analysis unit 200A performs multivariate analysis on the self-declared information and external evaluation information input as multiple explanatory variables, and calculates the matching rate (similarity and correlation) between the target person and the recruitment conditions in the feature vector space. As a result, the multivariate analysis unit 200A does not only focus on a certain explanatory variable to find the correlation and similarity between the target person and the dispatch destination, but can recommend dispatch destination candidates that take into account the subtle correlation and similarity in the high-dimensional feature space.

[0051] Figure 7 This is a diagram showing an example of a bidding calculation process in the embodiment.

[0052] The bid decision unit 230 may change the bid according to the bid request received by the bid receiving unit 130, and may change the bid fluctuation range according to the distance between the characteristic vector of the candidate and the characteristic vector of the recruitment conditions. The bid fluctuation range may, for example, represent the change of the bid updated in response to the increase in the number of bids for a certain candidate. The bid decision unit 230 of the talent value evaluation unit 214b may be the distance d( Figure 7 The closer the distance d between the feature vector of the target person and the feature vector of the recruitment condition is, the more the bid (y in the figure) is increased. The bid determination unit 230 increases the correction amplitude y for the initial bid g by multiplying the inverse of the distance d between the feature vector of the target person and the feature vector of the recruitment condition by logx, for example, as shown in the following formula.

[0053] y=g·xlogx(x≥1)org(x<1)

[0054] Thus, the bid determination unit 230 can change the bid increase more greatly for candidates whose recruitment conditions are closer to those of the dispatch destination business. On the other hand, the bid determination unit 230 can change the bid increase slightly when there are no candidates whose recruitment conditions are closer to those of the dispatch destination business.

[0055] Figure 8 1 is a sequence diagram showing an example of the operation process of the talent rating system 1 in the embodiment.

[0056] First, the candidate terminal device 300 sends registration information S10 including talent information to the talent bidding device 100. The talent information is sent to the talent rating device 200 and stored in the storage device 212. The dispatch source terminal device 400 sends registration information S12 including dispatch source information and candidate information to the talent bidding device 100. The dispatch source information and candidate information are sent to the talent rating device 200 and stored in the storage device 212. The dispatch source terminal device 400 may include fee information paid by the dispatch source business to the manager of the talent bidding device 100 for registering the candidate in the registration information S12. The dispatch destination terminal device 500 sends registration information S14 including dispatch destination information and recruitment information to the talent bidding device 100. The dispatch destination information and recruitment information are sent to the talent rating device 200 and stored in the storage device 212. The dispatch destination terminal device 500 may include fee information regularly paid by the dispatch destination business to the manager of the talent bidding device 100 in the registration information S14, and may include fee information for obtaining bid points. The talent bidding device 100 transmits fee information S16 requesting the dispatch destination business to pay the intermediary fee to the dispatch destination terminal device 500. The talent bidding device 100 transmits fee information S18 requesting the dispatch source business to pay the intermediary fee to the dispatch source terminal device 400. Thus, the operator of the talent bidding device 100 can obtain the intermediary fee from the dispatch destination business and the dispatch source business.

[0057] The talent bidding device 100 transmits point issuance information S20 indicating the bidding points issued by the issuing unit 110 to the dispatch destination merchant to the dispatch destination terminal device 500. The talent bidding device 100 transmits point issuance information S22 indicating the bidding points issued by the issuing unit 110 to the dispatch source merchant to the dispatch source terminal device 400.

[0058] The dispatch source terminal device 400 transmits the dispatch registration information S24 including the subject information of the dispatched subject among the registered subjects to the talent bidding device 100. The talent rating device 200 determines the rating based on the feature vector of the subject included in the dispatch registration information S24, and transmits the bidding information S26a indicating the bid to the talent bidding device 100 based on the rating information. The talent bidding device 100 transmits the rating information and the bidding information S26a to the subject terminal device 300, the dispatch source terminal device 400, and the dispatch destination terminal device 500. Thus, the subject, the dispatch source merchant, and the dispatch destination merchant can view the rating and the bid of the subject.

[0059] The talent rating device 200 uses the feature vector of the recruitment conditions and the feature vector of the candidate to extract the candidate having a feature vector close to the feature vector of the recruitment conditions, and sends matching information S26b including the candidate information of the extracted candidate to the talent bidding device 100. The information providing unit 120 sends the matching information S26b to the dispatch destination terminal device 500. Thus, the dispatch destination merchant can view the candidate matching the recruitment conditions.

[0060] The plurality of dispatch destination terminal devices 500 transmit bid request information S28a, 28b, ... including the subject information and bid point information of a subject to the talent bidding device 100. The bid decision unit 230 changes the bid price according to the received bid request. The dispatch destination decision unit 140 determines the dispatch destination merchant with the largest bid point among the plurality of bid request information S28a, 28b, ... as the successful bidder, and transmits successful bid information S30 indicating the successful bidder to the dispatch destination terminal device 500, the dispatch source terminal device 400, and the subject terminal device 300.

[0061] The talent bidding device 100 sends point information S32a obtained by subtracting the bid points from the bid points of the dispatch destination merchant that is the successful bidder to the dispatch destination terminal device 500 of the successful bidder, and sends point information S32b, ..., for returning the bid points to the dispatch destination merchants that did not win the bid, to the dispatch destination terminal devices 500 of the dispatch destination merchants other than the successful bidder. The issuing unit 110 of the talent bidding device 100 sends point information S34 for issuing the bid points as the success reward to the dispatch source terminal device 400 corresponding to the dispatch source merchant of the object of the successful bidder.

[0062] When the candidate works at the dispatch destination, the issuing unit 110 sends the cost information S36a indicating a part of the employment cost (hourly wage) as the labor cost of the candidate to the dispatch source terminal device 400 of the dispatch source merchant of the candidate who has won the bid. The cost information S36a may be information indicating the bid points. Thus, the dispatch source merchant can obtain compensation by dispatching the candidate. In addition, the talent bidding device 100 sends the cost information S36b indicating a part of the employment cost (hourly wage) to the dispatch destination terminal device 500 of the dispatch destination merchant. Thus, the dispatch destination merchant can obtain compensation by employing the candidate.

[0063] The talent checking device 104 may perform coordination screening processing to recommend candidates to the dispatch destination merchant. The talent checking device 104 may search for a dispatch destination merchant similar to the desired candidate based on the search history or bidding history of the dispatch destination merchant, and recommend a candidate similar to the candidate who won the bid at the searched dispatch destination merchant. The talent checking device 104 may recommend a candidate having a feature vector similar to the feature vector of the candidate viewed by the dispatch destination merchant by performing content-based screening processing. The talent checking device 104 may infer the feature vector of the recruitment conditions of the dispatch destination merchant through coordination screening processing, and recommend a candidate having a feature vector similar to the inferred feature vector of the recruitment conditions through content screening processing.

[0064] Furthermore, the learning device 214 can use the feature vector of the recruitment conditions of the dispatch destination merchant, the feature vector of the successful bidder of the dispatch destination merchant, and the successful bid (bid points) as learning data to learn the prediction model, input the feature vector of the recruitment conditions into the prediction model as an explanatory variable, and recommend the candidate having a feature vector similar to the feature vector of the candidate output from the prediction model and the initial value of the bid (inference result). The prediction model can use a statistical model such as a normal distribution and a binomial distribution, and has parameters for determining the statistical model. With respect to the parameters, the optimal value for outputting the inference result is set through the learning process.

[0065] In addition, the prediction model can be learned by any method in unsupervised learning and supervised learning. Unsupervised learning includes, for example, dimensionality compression method processing and clustering processing. Dimensional compression methods include, for example, principal component analysis, multidimensional scaling method, t-SNE method, etc. Clustering processing includes k-means method, hierarchical clustering, etc. Supervised learning includes, for example, classification processing or regression processing, etc. Classification processing includes, for example, decision tree, support vector machine, random forest, logistic regression, etc. Regression processing includes, for example, partial least squares (Partial Least Squares Regression, PLS), Lasso regression (least absolute shrinkage and selection operator, LASSO), Ridge regression, support vector machine, random forest, logistic regression, etc. The learning device 214 uses these learning methods to construct a prediction model, and uses the constructed prediction model to perform processing in the multivariate analysis unit 200A (operation unit 210, rating determination unit 220, and bidding determination unit 230).

[0066] As described above, according to the implementation method, a talent rating system 1 can be realized, which includes: a subject terminal device 300; an external device 600 for evaluating the subject; and a talent rating device 200, to which the subject terminal device 300 and the external device 600 are connected via a communication network NW, wherein the talent rating device includes: an operation unit 210, which inputs the talent information of the subject obtained from the subject terminal device 300 and the evaluation result of the subject obtained from the external device 600, and outputs the characteristic information of the subject based on the talent information and the evaluation result; and a rating determination unit 220, which determines the rating of the subject based on the characteristic information output from the operation unit 210.

[0067] According to the talent rating system 1, the subject can be rated based on the information reported by the subject as the talent information of the subject and the evaluation result of the subject obtained by the external organization as the evaluation result, so that the talent can be comprehensively evaluated and the rating of the subject and the dispatch destination business can be visualized.

[0068] According to the talent rating system 1, the bid of the target person can be determined based on the rating determined by the rating determination unit 220, the bidding information indicating the bid determined by the bidding determination unit 230 is sent to the dispatch destination terminal device 500, the dispatch destination terminal device 500 receives a bidding request including the target person information and the bidding point information, and the dispatch destination of the target person is determined based on the bidding request. Thus, according to the talent rating system 1, the bidding to the target person can be supported based on the rating.

[0069] According to the talent rating system 1, for example, when a licensed pharmacist is envisioned as the subject, by winning the bid for a licensed pharmacist who can work in a holiday clinic or a pharmacy in an area where there is a shortage of licensed pharmacists, it is possible to suppress the scenario of a shortage of licensed pharmacists and provide medicines to patients. In addition, according to the talent rating system 1, there are the following advantages: in order to improve the skills of licensed pharmacists, companies and pharmacies can dispatch licensed pharmacists to workplaces with different types of business or working methods as part of education, and for the dispatch source merchants, the skills of the licensed pharmacists they have can be improved, which can increase the bids of licensed pharmacists. Moreover, according to the talent rating system 1, training content can be provided free of charge in response to the information of the registered subject, and the bids of licensed pharmacists can be increased based on the attendance history of the training. Therefore, according to the talent rating system 1, it is possible to visualize the bids corresponding to the skills of the licensed pharmacists themselves, and it is possible to increase the value seen by the dispatch destination merchants.

[0070] According to the talent rating system 1, a feature vector based on talent information and the evaluation result can be calculated, the rating can be changed according to the calculated feature vector, and the initial value of the bid can be changed according to the determined rating. Thus, the talent rating system 1 can set a bid for evaluating the subject appropriately according to the subject's qualifications and evaluation.

[0071] According to the talent rating system 1, a feature vector can be calculated based on the recruitment conditions received from the dispatch destination terminal device 500, and matching information including information of the target person having a feature vector and a feature vector close to the feature vector of the recruitment conditions can be sent to the dispatch destination terminal device 500. Thus, the talent rating system 1 can save the time of manually searching for a target person close to the recruitment conditions.

[0072] According to the talent rating system 1, the dimensions of the feature vector of the subject and the feature vector of the recruitment conditions can be compressed, and the subject having a feature vector close to the feature vector of the recruitment conditions can be extracted under the compressed dimension, so the process of extracting the subject close to the recruitment conditions can be accelerated.

[0073] According to the talent rating system 1 , the bid price can be changed according to the bid request, and the fluctuation range of the bid price can be changed according to the distance between the feature vector of the candidate and the feature vector of the recruitment conditions.

[0074] In addition, although each embodiment and variation example are described, this is only an example and is not limited to these. For example, any scheme in each embodiment or variation example, a part of each embodiment, a part of each variation example may be combined with one or more other embodiments or one or more other variation examples to implement a method of the present invention.

[0075] [Explanation of Symbols]

[0076] 1: Talent bidding support system; 100: Talent bidding device; 102: Point management device; 104: Talent verification device; 110: Issuing department; 112: Consumption department; 120: Information provision department; 130: Bid acceptance department; 140: Dispatch destination determination department; 200: Talent rating device; 200A: Multivariate analysis department; 202: Talent registration device; 204: Enterprise demand registration department; 206: Talent information registration department; 210: Calculation department; 212: Storage device; 212a: Member information Database; 212b: database for learning; 214: learning device; 214: talent characteristic evaluation unit; 214a: talent characteristic evaluation unit; 214b: talent value evaluation unit; 214c: learning information update unit; 220: rating decision unit; 230: bidding decision unit; 300: subject terminal device; 400: dispatch source terminal device; 410: comprehensive evaluation unit; 500: dispatch destination terminal device; 600: external device; 700: training content providing device; 710: examination implementation unit.

Claims

1. A talent rating system having: The subject terminal device used by the subject; an external device for evaluating the subject; and A talent rating device is connected to the subject terminal device and the external device via a communication network. The talent rating device comprises: a calculation unit that inputs the talent information of the subject obtained from the subject terminal device and the evaluation result of the subject obtained from the external device, and outputs the characteristic information of the subject based on the talent information and the evaluation result; and The ranking determination unit determines the ranking of the subject based on the feature information output from the calculation unit.

2. The talent rating system according to claim 1, wherein: The talent information of the subject is information reported by the subject, and the evaluation result is a result of evaluation of the subject by an external organization.

3. The talent rating system according to claim 1, wherein: The talent rating system has: A terminal device at a dispatch destination used by a business to which the talent is dispatched; and The talent bidding device is connected to the target terminal device and the dispatch destination terminal device via a communication network. The talent rating device includes a bid determination unit that determines a bid of a target person based on the rating determined by the rating determination unit. The talent bidding device comprises: an information providing unit that transmits bid information indicating the bid determined by the bid determining unit to the dispatch destination terminal device; a bid accepting unit that accepts a bid request including target person information and bid point information from the dispatch destination terminal device; as well as The dispatch destination determination unit determines a dispatch destination of the target person based on the bid request.

4. The talent rating system according to claim 3, wherein: The calculation unit calculates a feature vector based on the talent information and the evaluation result. The ranking determination unit changes the ranking according to the feature vector calculated by the calculation unit, The bid determination unit changes an initial value of the bid according to the rating determined by the rating determination unit.

5. The talent rating system according to claim 4, wherein: The calculation unit calculates a feature vector based on the recruitment conditions received from the dispatch destination terminal device. The information providing unit transmits matching information including information of the target person having the feature vector of the target person calculated by the calculation unit and a feature vector close to the feature vector of the recruitment condition calculated by the calculation unit to the dispatch destination terminal device.

6. The talent rating system according to claim 5, wherein: The calculation unit compresses the dimensions of the feature vector of the subject person and the feature vector of the recruitment condition, and extracts the subject person having a feature vector close to the feature vector of the recruitment condition in the compressed dimension.

7. The talent rating system according to claim 5, wherein: The bid determination unit changes the bid in accordance with the bid request accepted by the bid acceptance unit, and changes a fluctuation range of the bid in accordance with a distance between a feature vector of the target person and a feature vector of the recruitment condition.

8. A talent rating method, which is a talent rating method of a talent rating system, the talent rating system comprising: a subject terminal device used by a subject; an external device for evaluating the subject; and a talent rating device connected to the subject terminal device and the external device via a communication network, wherein: The talent rating methodologies include: The subject terminal device sends the subject's talent information to the talent rating device; The external device sends the evaluation result of the subject to the talent rating device; The talent rating device acquires talent information of the subject and evaluation results of the subject; The talent rating device calculates characteristic information of the subject based on the talent information and the evaluation result; and The talent rating device determines the rating of the subject based on the characteristic information.

9. A talent rating device, connected to a subject terminal device used by a subject and an external device for evaluating the subject via a communication network, wherein: The talent rating device comprises: a calculation unit that inputs the talent information of the subject obtained from the subject terminal device and the evaluation result of the subject obtained from the external device, and outputs the characteristic information of the subject based on the talent information and the evaluation result; as well as The ranking determination unit determines the ranking of the subject based on the feature information output from the calculation unit.

10. A talent rating program, wherein a computer of a talent rating device connected to a subject terminal device used by a subject and an external device for evaluating the subject via a communication network functions as the following means: a calculation unit that inputs the talent information of the subject obtained from the subject terminal device and the evaluation result of the subject obtained from the external device, and outputs the characteristic information of the subject based on the talent information and the evaluation result; and The ranking determination unit determines the ranking of the subject based on the feature information output from the calculation unit.

Images